performance effects of venture capital firm networks · performance effects of venture capital firm...
TRANSCRIPT
1
Performance Effects of Venture Capital Firm Networks
Peter Abell Department of Management
London School of Economics Houghton Street
London WC2A 2AE UK Email: [email protected]
Tahir M. Nisar School of Management
University of Southampton, Highfield, Southampton SO17 1BJ UK
Email: [email protected]
2
Performance Effects of Venture Capital Firm Networks
Abstract: This paper is primarily concerned with the question of the networking effects of venture capital (VC) firms on portfolio companies. VCs can bring specific skills and abilities to their on-going relationships with their portfolio companies and thus add value by influencing key operations. High levels of engagement also translate into giving advice and support, helping with the team culture, creating strategic alliances, or exercising corporate governance. A particular mechanism through which these support services are delivered is syndication investment. Using network theory tools we investigate the effects of syndication on VC firm performance. We find that better-networked VC firms experience significantly better performance, as measured by the proportion of investments that are successfully exited through an IPO (Initial Public Offering) or a sale to another company. Keywords: Networking; Syndicates; Value-adding activities; Venture capital.
Introduction
Networks may be seen as an alternative to banks as intermediaries in credit markets.
They may also coordinate disperse resources in factor and product markets. Networks’
role in the allocation of scarce resources has been recognized both in the past and today
(Cornelli and Goldreich, 2001; Ljungqvist, Marston, and Wilhelm, 2005). As an
organizational form a network is distinct from both a firm and a market, as they are not
limited by specific contracts to some ultimate owner of property rights, nor are they
engaged in a bargaining process, having to bid for resources. Networks are rather
cooperative mechanisms that arrive at allocative decisions by consensus and through the
pooling of relevant information.
In the venture capital (VC) industry, networks abound. A central feature of venture
capital is the way it allows investment firms to syndicate their investments- when two or
more firms jointly undertake a new project (Lerner, 1994). Syndication inevitably creates
a multitude of relationships spanning the syndicate members coordination and various
service providers, including research & development organizations, patent lawyers, head
hunters, investment bankers etc (Gorman and Sahlman,1989; Sahlman, 1990). The
diverse sets of relationships with different agents (e.g., investors, co-investors, suppliers,
customers), imply large variations in the quality and support offered to portfolio
3
companies. It is likely that these differences in investment and engagement approach will
also affect VC syndicate performance.
VC firms operate in uncertain environments. The uncertainties are often linked to the risk
involved in product development and commercialization. For instance, risks faced by
start-up companies have their roots in both the external environment and in lack of
capabilities within the organization. There may be sudden market changes and shifts in
investor and consumer expectations, compelling the company to completely change the
course of its strategy. Companies may also face shortage of personnel or key physical and
organizational resources to carry out a new development opportunity successfully. VC
literature has argued that by syndicating investments companies are able to effectively
deal with all such risks and uncertainties.
Prior literature on networks has primarily been concerned with identifying their existence
in various financial markets. An important issue is the likely impact of this particular
form of organizational structure on VC firm performance. To this end, VC network
measures of how well networked a VC is will need to be developed. Recently, there has
been a surge of interest in networking theory, which freely borrows its main tools of
analysis from Graph theory (Hochberg, Ljungqvist and Lu, 2006). In the present context,
we use the graph theory based measurement tools to capture the following different
aspects of a VC firm’s influence: (i) the number of VCs with which it has a relationship
as a proxy for information about deal flow, expertise, contacts, and pools of capital it has
access to; (ii) its access to the best-connected VCs; and (iiii) its ability to act as an
intermediary, bringing together VCs with complementary skills or investment
opportunities that lack a direct relationship. As the prevailing practice of syndication
investment underlines VC firms greatly benefit from having a multitudes of relationships.
Our results show that the benefits from these relationships are magnified when VC firms
join hands with other well-connected VC firms.
The paper is organized as follows. The next section provides a survey of the prior
literature on VC firm engagement practices. This section also develops VC specific
4
network theory measures. We then discuss our data and provide variable definitions,
which is followed by our report of results. We establish empirical implications for the
system of VC networked relations, including the impact of various measures of networks
on VC firm performance. In the final section, we conclude and discuss the research’s
implications.
Literature Survey
The evolution of interdependencies between different organizational tasks is a
manifestation of how linkages between different actors play a critical role in enterprise
development. To better understand the increasingly complex relationships among
organizations, which often arise, as we pointed out above, as a result of their uncertain
market environments (Kenis and Knoke, 2002; Provan and Sebastian, 1998), researchers
have often resorted to analysing resources and capabilities as catalysts for
interorganizational relations. Such an approach is often couched as the resource-based
view of the firm (Barney, 1991; Wernerfelt, 1984). Although this approach recognizes
the role of unique firm history and specific nature of capabilities, it pays scant attention to
network ‘complementarities’ that motivate firms’ participation in interfirm networks. It is
important to note that interfirm networks can be useful in this framework as they channel
valuable information about the resource advantage (e.g., key information, useful contacts,
capital) that can be obtained from firm network resources.
Prior literature on syndication has generally focused on ex-anti motives of VC firms,
namely how syndication may improve the selection process through improved screening,
due diligence and decision making. Following traditional finance theory, syndication is
considered as a means by which VC firms secure a well-diversified portfolio, enabling
the VC firm to reduce the unsystematic risk of the portfolio. There are also expectations
of reciprocity and better future financial deals when VC firms are able to syndicate with
other reputable firms.
5
For a number of reasons, determining the ex-post motives of syndication, especially how
investors can share their complementary skills or specific knowledge, and as a result, add
value to a portfolio company is also important (Bygrave, 1988). The management of the
ex-post investment may significantly affect its performance. Brander et al. (2002)
compared Canadian-syndicated VC deals with stand-alone projects and found that the
need to access specific resources from the ex-post management of investment, rather than
for the selection of investment resulted in the syndication of projects which also enjoyed
higher rate of returns than stand-alone projects. The study of such investments provides
an opportunity to examine the key factors driving the VC firm performance.
Network Methodology
Network analysis is conducted at various levels of social interaction. One can either
analyse clusters (networks) of organizations and their patterns of interaction, involving a
study of the complete network or focus solely upon one or more specific operations in a
network (Powell, 1990; Burt, 1992; Powell et al., 1996). Within the field of business
studies, one common practice is to take the firm as a unit of analysis and examine the
links or ties it is able to build with actors outside the firm. This helps to understand how
the firm develops and operates a network (Birley, 1985; Aldrich and Zimmer, 1986;
Dubini and Aldrich, 1991; Larson, 1992). Notwithstanding the fact that counting the
number of ties a firm has established provides important insights into its network size, the
various dimensions of a network and/or the way they influence the development of a new
business is also a key question in network analysis (Chu, 1996). For example, questions
such as how socio-economic contexts in which actors operate are linked to strategic
networks have not yet been fully explored.
The social network model of organization formation (Granovetter, 1973, 1985; Aldrich
and Zimmer, 1986; Dubini and Aldrich, 1991; Burt, 1992; Larson, 1992) views new
organizations as created by entrepreneurs ‘embedded’ (Granovetter, 1985) in a series of
network relationships. Specifically, the pattern and process through which network
relationships are developed and maintained are the main areas of research interest. Dubini
6
and Aldrich (1991) define a network as ‘patterned relationships between individuals,
groups, and organizations’. This suggests that a network consists of structure and content
(Burt, 1992), where structure is the pattern by which individuals and organisations are
linked, including interaction frequency and the ways in which they are connected (e.g. by
contracts, equity, kinship, etc.), while content describes the resources exchanged within
networks.
Network size
Extant literature examines the structure of organizational networks in terms of size and
strength (Dubini and Aldrich,1991). There are many ways in which network size is
analysed. A common approach is to study the number of ties or links between an
organization and outside contacts. In this respect, one can usefully distinguish between a
‘personal’ network, i.e. ‘those persons with whom an entrepreneur has direct relations
and an ‘extended’ network, i.e. relationships between an organization and the external
world (Dubini and Aldrich, 1991). A related distinction is between the actor’s direct
‘personal’ links and ‘total’ links which cover direct and indirect links – Shulman’s
‘anchorage’ points within the focal organization (1976). Shulman emphasizes the
importance of both direct and indirect links as they are used as a means of accessing
resource nodes. For example, Steier and Greenwood (1995) show how venture capitalists
act as brokers, linking entrepreneurial ventures with such players as retail outlets,
marketing consultants and other financial sponsors. In this line of research, the main unit
of analysis is the number of links between an organization and its context. Irrespective of
whether the links are direct or indirect, what matter is the number of links (Larson, 1992).
Another strand of research emphasizes the number of links to different clusters of
information and resources accessed by those resources, rather than the number of ties per
se (Burt, 1992). ‘Size is a mixed blessing. More contacts can mean more exposure to
valuable information, more likely early exposure, and more referrals. But increasing
network size without considering diversity can cripple the network (Burt, 1992). Burt
stresses the role of diversity as a key attribute of size, implying that several single ties to
7
multiple sources - ties to clusters of people not in communication with other clusters -
than multiple ties to the same source, are more important.
The number of relationships an actor in a network manages is captured by degree
centrality measure. For research purposes, a network is represented by a square
“adjacency” matrix, the cells of which reflect the ties among the actors in the network.
For example, two VCs coinvesting in the same portfolio company can be considered as
having a tie. With the number of ties increasing, the VCs would have more opportunities
for exchange and so become influential, or central, in a network. We examine this by
looking at whether at least one syndication relationship exists between VCs i and j.
Adjacency matrices can be “directed” or “undirected”. In the case of directed matrices,
we can differentiate between the originator and the receiver of a tie. Thus, we can have
information about syndicates led by VC i versus those led by VC j. It allows one to
distinguish between VCs who receive many ties and those who originate many ties. In the
first case, VCs are invited to be syndicate members by many lead VCs, whilst, in the
other case, lead VCs syndicate with many other VC members. This produces two directed
measures of degree centrality. VC i’s indegree counts if at least one syndication
relationship exists in which VC j is the lead investor and VC i is a syndicate member.
When one counts the number of other VCs a VC firm has invited into its own syndicates
(i.e., reciprocity), one has an outdegree measure.
Network strength
Granovetter (1973, 1985) argues that network ‘strength’ where ties with close friends and
associates who are likely to know one another, is critical for understanding the network
relations. Whilst recognizing the value in weak ties as they represent diverse conduits of
information, this framework focuses on ‘strong’ ties as, for example, relations within
personal networks, that allow trust and/or moral obligation to play its role as an
underlying support mechanism, thus reducing the risk of opportunism and cheating
(Johanson and Mattson, 1987; Powell, 1990). Therefore, in addition to counting the
number of actor relationships (i.e., degree), it would also be useful to measure their
8
‘strength’ or ‘quality’ or closeness in our setting. “Eigenvector centrality” is generally
used to measure “closeness” (Bonacich, 1987), which is an actor’s total ties to other
actors weighted by their respective centralities. Thus, we can measure the extent to which
a VC is connected to other well connected VCs.
Network content
In addition to structure, understanding network content is also important in the way it
affects individual actors and organizations. In this respect, Tichy (1981) develops a
typology of four types of resources accessed through networks: information, goods and
services, expressions of affect or emotional support, and political influence. Information
resources can be envisaged in terms of management expertise, and professional advice
provided by accountants, lawyers and consultants. In the present context, goods and
services are financial resources, as well as assets such as plants and equipment. Steier and
Greenwood (1995) discuss effective support in relation to its significance for start-up
companies. They also provide an example of how networks can be used for political
purposes or for the mobilisation of influence. Thus, it will be useful to know those actors
on whom many others must rely to make connections within the network. When a VC
acts as an intermediary by bringing together VCs with complementary skills that lack a
direct relationship between them, this can be captured by ‘betweenness’ measure. It is
measured by the proportion of all paths linking actors j and m that pass through actor i.
Indirectly, these network measures also capture the involvement of VC firms in portfolio
companies. Elango, Fried, Hisrich and Polonchek (1995) identify three levels of
involvement: inactive, active advice-giver, and hands-on. Involvement by the inactive
group is mainly confined to attendance at board meeting. One example of ‘hands-on’
involvement approach is the firms’ emphasis on networking among its portfolio
companies. The concept refers to networks of companies bound together by mutual
obligations and contacts. Entrepreneurs gain access to VC’s portfolio of companies and
associations with market leaders. These relationships are the foundations for strategic
alliances, partnership opportunities and the sharing of insights to help build new ventures
9
faster, broader and with less risk (Norton and Tenenbaum, 1993). Networking also allows
the development of a network of companies that helps create synergies and pooled
resources for growth and development (Franke et al., 2006). The professional respect and
regard in which VC is generally held and the influence that it enjoys with other investors
are major sources of networking success.
Empirical Specifications
VCs often engage in joint projects that allow them to participate in more projects,
especially when they are constrained by available resources. There is empirical evidence
that suggests that syndicated projects offer higher returns than projects financed by a
single VC (stand-alone projects) (Brander et al., 2002). The way syndicated VCs offer
improved managerial support in the form of higher reputation, and a larger variety of
contacts for their portfolio firms than a single VC, accounts for the higher return. The
syndication investment strategy is primarily aimed at high risk environments, where it
reduces the asymmetries of information and select the projects of the highest quality
(Wilson, 1968). By combining equities in an investment, either in the same investment
round or, more broadly defined, at different points in time, syndicated projects mitigate
investment risks.
Syndication may also be advantageous when inexperienced VCs join in. Experienced
VCs are likely to contribute to the project’s success with their knowledge and expertise,
and thus increase the expected project value. In the process, they are also likely to help
inexperienced VCs to gain valuable know-how for future deals when they invest in a
project together with skilled partners. To fully investigate the effects of syndication
investment, we examine directed and undirected centrality measures as discussed above
(see also Hochberg, Ljungqvist and Lu, 2006). Entry to and exit from the network may
change each VC’s centrality, thus changing the structure of the network and the
relationships it supports. To keep track of these changes, we construct our adjacency
matrices over trailing three-year windows.
10
Descriptive statistics are presented in Table 1. The parent of the average sample fund has
normalized outdegree of 8.13%, indegree of 7.12%, and degree of 12.14%. This suggests
a number of interesting features of syndication investment in the UK and Continental
Europe: the average VC, when acting as lead, involves a little over 8% of all VCs active
in the market at the time as coinvestors; has been invited to become a syndicate member
by around 7% of all VCs; and has coinvestment relationships with a little over 12% of the
other VCs (ignoring its and their roles in the syndicate). These relatively low degree
centrality scores suggest that VC relationships are relatively exclusive and stable as they
often coinvest with a small set of other VCs. There is also the existence of a special class
of VCs who as a policy do not syndicate their investments.
Variables Definition
The sample consists of 624 venture capital funds based in the U.K and Continental
Europe. We examine the period between 1995 and 2005. The classification into seed or
early-stage funds follows Venture Economics’ fund focus variable. Corporate VCs are
identified manually starting with Venture Economics’ firm type variable. We augmented
the Venture Economics data with information from FAME, VentureExpert, and Diane.
Exit rate: percentage of portfolio companies exited
IPO rate: percentage of portfolio companies sold via IPO
M&A rate: percentage of portfolio companies sold via M&A
Dollar exit rate: percentage of invested $ exited.
Dollar IPO rate: percentage of invested $ exited via IPO
Dollar M&A rate: percentage of invested $ exited via M&A.
B/M is the book/market ratio of public companies in the sample fund’s industry of
interest.
Size is the amount of committed capital reported by a VC fund.
11
VC experience is defined as the average number of years of VC firm’s experience in
venture capital industry.
Partner experience is defined as the average number of years of partners’ experience in
venture capital industry.
Corporate board is a dummy variable that takes the value 1 if the venture capital firm is
reported to sit or have sat on the board of directors of the company; 0 otherwise.
Each of the following network measure is normalized by the theoretical maximum (e.g.,
the degree of a VC who has syndicated with every other VC in the network).
Degree is defined as the number of unique VCs a firm has syndicated with (regardless of
syndicate role).
Indegree is defined as the number of unique VCs that have led syndicates the firm was a
non-lead member of.
Outdegree is defined as the number of unique VCs that have taken part as non-lead
investors in syndicates led by the firm.
Eigenvector is a variable that measures ‘closeness’ in terms of how close to all other
VCs a given VC is.
Betweenness is defined as the number of shortest distance paths between other VCs in
the network with which the VC interacts.
Industry is a set of dummy variables that we obtain from our survey instrument, which
gave the following choices: Biotech and pharma; Medical products; Software and
internet; Financial services; Industrial services; Electronics; Consumer services;
12
Telecom; Food and consumer goods; Industrial products (incl. energy); Media &
Entertainment; Other (specify).
Table 1: Descriptive Statistics
Mean Median Min Max No.
Fund
characteristics
-fund size ($m)
-first fund
(fraction, %)
-seed or early-
stage fund
(fraction, %)
- corporate VC
(fraction, %)
34.0
13.2
18.7
8.3
11.0
0.1
2,700
624
Fund
performance
-exit rate (% of
portfolio
companies
exited)
- IPO rate (%
of portfolio
companies sold
via IPO)
- M&A rate (%
of portfolio
companies sold
via M&A)
- dollar exit
rate (% of
16.7
9.2
6.3
17.1
17.0
7.8
4.7
14.6
0
0
0
0
100
100
100
100
624
624
624
624
13
invested $
exited)
- dollar IPO
rate (% of
invested $
exited via IPO)
- dollar M&A
rate (% of
invested $
exited via
M&A)
12.3
7.2
6.8
2.4
0
0
100
100
624
624
Investment
climate
-average B/M
ratio
0.273
0.258
0.09
0.873
624
VC
experience
7.5
4
1
45
624
Partner
experience
11.34
14
0
25
624
Corporate
board
0.622
-
0
1
621
Network
measures
-Outdegree
-Indegree
-Degree
-Betweenness
-Eigenvector
8.13
7.12
12.14
0.311
4.86
6.24
5.20
10.32
0.11
2.74
0
0
0
0
0
24.2
14.2
45.17
10.87
42.53
624
624
624
624
624
14
Benchmark Analysis
Earlier studies on venture capital and VC fund performance have often found the
significant impact of log fund size and a set of vintage year dummies on performance
(Kaplan and Schoar, 2005; Hochberg, Ljungqvist and Lu, 2006). Table 2 presents a
similar type analysis.
In the first Column, we report results for when only size is included. In this case, we do
not find any significant relationship. We then use a dummy that equals one for first-time
funds in Column 2. Assuming that seed or early-stage funds invest in riskier companies
and thus have relatively fewer successful exits, we also include in this Column a control
for such investment vehicles. Similarly, we control for corporate VCs. As expected, first-
time funds perform significantly poorly. The results for the seed and early stage funds are
similar, but the coefficients on corporate VCs are positive. Our subsequent analysis
(Column 3) includes a proxy for the investment opportunities available to funds when
deploying their committed capital (using B/M ratio). In this case, we observe the impact
of a more favourable investment climate at the time a fund invested its capital as
significantly higher exit rates are recorded thereafter.
The right of venture firms to nominate directors of portfolio companies is an important
component of their control rights. The role of VC firm nominees may, however, extend
beyond control. Goodstein, Gautman and Boeker (1994) identify three functional duties
of company boards: (1) networking activities which are to do with forming links between
the company and its external environment, and securing critical resources; (2) monitoring
activities which include dealing with internal governance issues, monitoring company
performance and providing mechanisms to align the interests of management with
shareholders; and (3) strategy-making activities contributing to the company’s strategic
decision-making processes. Wijbenga, Postma, van Witteloostuijn and Zwart (2003)
follow this framework in reviewing the role of private equity investors in the boards of
portfolio companies.
15
First, in relation to networking activities, they suggest that there is evidence that the
boards of portfolio companies (on which general partners usually sit) provide an interface
with other investor group members, provide industry contacts and contacts with
professionals. Private equity firms also assist in obtaining alternative equity capital.
Second, in relation to monitoring activities, there is evidence that the boards of portfolio
companies monitor financial performance, monitor operational performance, and evaluate
the portfolio company’s strategy and product market opportunities in order to develop the
new venture’s strategy to changing circumstances. Finally, in relation to strategy-making
activities, there is evidence that the boards of portfolio companies serve as a sounding
board, assist in formulating business strategy, assist in dealing with short-term crises or
problems, and recruit and/or replace managers.
To investigate these matters, in Column 4 our variables of interest are the venture and
investment experience of the fund’s parent firm and firm’s partners. We find a
significantly improved performance for funds with more experienced parents and
individual partners. We find similar results for portfolio company board membership,
thus emphasizing the literature’s conclusions, as discussed above.
Table 2: Benchmark Analysis (ordinary least-squares regressions)
(1) (2) (3) (4)
Fund
characteristics
In fund size 0.031
(0.011)
0.053***
(0.012)
0.042***
(0.011)
0.039***
(0.011)
In fund size
squared
-0.003
(0.001)
-0.006***
(0.001)
-0.003**
(0.002)
-0.004***
(0.001)
=1 if first fund -0.067***
(0.017)
-0.041***
(0.012)
0.007
(0.012)
=1 if seed or
early-stage
fund
-0.053***
(0.013)
-0.016***
(0.009)
-0.024**
(0.009)
16
=1 if corporate
VC
0.048**
(0.011)
0.031
(0.016)
0.033
(0.016)
Investment
climate
-average B/M
ratio
-0.423***
(0.012)
-0.276***
(0.019)
VC
experience
0.033***
(0.015)
Partner
experience
0.031***
(0.010)
Corporate
board
0.024***
(0.009)
-Adjusted-R2
-Test: all
coefficients =
0 (F)
-No. of
observations
18.2%
41.6***
621
16.1%
32.5***
621
19.4%
47.9***
621
24.3%
33.7***
621
Notes: The dependent variable is a firm’s exit rate, as defined above. Year dummies
controlling for vintage year effects are included but not reported. White
heteroskedasticityconsistent standard errors are shown in italics. ***, **, and * denote
significance at the 1%, 5%, and 10% level.
VC Networks and Performance
In the previous section, we investigated a benchmark VC firm performance model, while
controlling for fund characteristics, competition for deal flow, investment opportunities,
corporate board and venture and parent firm experience. Now, we investigate the extent
to which VC’s networking improve the performance of its fund. The results are largely
consistent with the predictions of our framework. In Table 3, we add our five network
measures to the specifications shown in Table 2, thus presenting estimates for four
17
separate regression models. We find that networked VC firms are associated with
significantly better fund performance in each of the specification.
The results show that there are three significant network effects; indegree stands out as
the most important factor in relation to its effect on network relations. This is closely
followed by two other measures – eigenvector and degree. Overall, these results suggest
that large benefits accrue to VC firms when they have many ties (degree). These ties have
more store when VCs are invited into many syndicates (indegree), and when they involve
other well-connected VCs (eigenvector). Another measure, outdegree, has a similarly
relatively large effect, which implies that a VC firm’s investment in future reciprocity
pays off handsomely (see also Lerner, 1994). When we examine the significance of a VC
firm’s ability to act as an agent between other VCs (betweenness) we do not find a
significant relationship. This reveals the fact that indirect relationships (those requiring
intermediation) are much less important in the way venture capital industry is organized.
From these results, one can also argue that networked firms are best placed to adopt a
hands-on approach to managing their portfolio companies. As discussed, VC firms taking
a hands-on approach both monitor portfolio firms through, for example, reviewing
management accounts and board minutes, and through involvement in decisions such as
the purchase of major capital items, acquisitions and disposals, changes in strategic
direction, appointment of directors and auditors, and changes in capital structure. The
alternative hands-off, or passive approach, mainly involves monitoring portfolio firms,
for example through management accounts. VC firms taking a hands-off approach are,
however, likely to become actively involved with the portfolio company under certain
circumstances such as failure to meet agreed targets or default of payments.
Table 3: VC Networks and Performance (ordinary least-squares regressions)
(1) (2) (3) (4)
Fund Characteristics
-In firm size 0.044***
(0.016)
0.041***
(0.016)
0.042***
(0.016)
0.039***
(0.016)
18
-In firm size squared -0.004**
(0.001)
-0.005**
(0.001)
-0.004**
(0.001)
-0.004**
(0.001)
=1 if first fund 0.008
(0.013)
0.008
(0.013)
0.008
(0.013)
0.011
(0.013)
=1 if seed or early-stage
fund
-0.035**
(0.011)
-0.036**
(0.011)
-0.028**
(0.011)
-0.036**
(0.011)
=1 if corporate VC 0.042
(0.017)
0.039
(0.017)
0.034
(0.017)
0.032
(0.017)
Investment climate
-average B/M ratio
-0.411***
(0.045)
-0.414***
(0.045)
-0.408***
(0.045)
-0.404***
(0.045)
VC experience 0.011**
(0.009)
0.013***
(0.009)
0.011**
(0.009)
0.013**
(0.009)
Partner experience 0.017**
(0.003)
0.014**
(0.002)
0.015**
(0.001)
0.014**
(0.002)
Corporate board 0.007***
(0.001)
0.009***
(0.001)
0.008***
(0.001)
0.010***
(0.001)
Network measures
-Degree
-Indegree
-Outdegree
-Eigenvector
-Betweenness
0.016***
(0.008)
0.019***
(0.002)
0.017***
(0.003)
0.014***
(0.002)
0.009***
0.014***
(0.008)
0.021***
(0.009)
0.016***
(0.003)
0.016***
(0.005)
0.007***
0.018***
(0.005)
0.021***
(0.009)
0.017***
(0.003)
0.016***
(0.005)
0.007***
0.016***
(0.008)
0.019***
(0.002)
0.019***
(0.007)
0.014***
(0.002)
0.007***
19
(0.002) (0.001) (0.001) (0.001)
-adjusted-R2
-Test: all
coefficients = 0 (F)
-No. of observations
22.4%
54.2***
621
23.8%
56.6***
621
21.6%
53.4***
621
25.7%
55.1***
621
Notes: Notes: The dependent variable is a firm’s exit rate, as defined above. Year
dummies controlling for vintage year effects are included but not reported. White
heteroskedasticityconsistent standard errors are shown in italics. ***, **, and * denote
significance at the 1%, 5%, and 10% level.
In our analysis, we find that the better-networked VCs generate rents by developing
superior relational approaches. For example, when a VC in a network reports to the other
members about the accomplishment of pre-set performance targets, both partners generate
rents through exchange of knowledge valuable to one another. Frequent interactions
between the parties may enhance access to each other’s knowledge base and increase the
capability of processing complex knowledge. However, this analysis does not take into
account the possibility that this networking approach is valued only when it affects
performance. To account for the endogenous relation between VC networks and
performance, we also estimate the performance effects of VC network characteristics
with two stage least squares (2SLS). Table 4 provides 2SLS results, which show that VC
networks indeed raise firm performance. In fact, the results are stronger than the OLS
regressions, especially the association between degree and VC performance.
Table 4: Endogeneity: The Effects of VC Networks (2SLS)
(1) (2) (3) (4)
Firm
Characteristics
In Firm size 0.042***
(0.014
0.045***
(0.014)
0.046***
(0.014)
0.048***
(0.016)
In Firm size squared -0.006**
(0.003)
-0.007**
(0.003)
-0.007**
(0.003)
-0.007**
(0.003)
20
=1 if first fund 0.009
(0.013)
0.007
(0.013)
0.006
(0.013)
0.006
(0.013)
=1 if seed or early-
stage fund
-0.028**
(0.011)
-0.021*
(0.011)
-0.021*
(0.011)
-0.021*
(0.011)
=1 if corporate VC 0.023
(0.017)
0.021
(0.017)
0.022
(0.017)
0.023
(0.017)
Investment climate
-average B/M ratio
-0.413***
(0.052)
-0.386***
(0.052)
-0.374***
(0.052)
-0.422**
(0.052)
VC experience 0.014***
(0.011)
0.011***
(0.010)
0.014***
(0.011)
0.016***
(0.013)
Partner experience 0.004***
(0.001)
0.005***
(0.001)
0.004***
(0.001)
0.005***
(0.001)
Corporate board 0.018***
(0.005)
0.016***
(0.005)
0.015***
(0.005)
0.018***
(0.005)
Network measures
-Degree
-Indegree
-Outdegree
-Eigenvector
-Betweenness
0.023***
(0.014)
0.026***
(0.017)
0.019**
(0.008)
0.015***
(0.011)
0.011***
(0.002)
0.024***
(0.016)
0.025***
(0.011)
0.021**
(0.008)
0.015***
(0.011)
0.011***
(0.002)
0.024***
(0.016)
0.026***
(0.017)
0.019**
(0.008)
0.017***
(0.012)
0.011***
(0.002)
0.023***
(0.014)
0.025***
(0.011)
0.021**
(0.008)
0.015***
(0.011)
0.011***
(0.002)
21
-adjusted-R2
-Test: all
coefficients=0 (F)
-No of observations
21.1%
51.3***
621
23.4%
52.4***
621
25.8%
53.4***
621
27.1%
55.1***
621
Notes: The dependent variable is a firm’s exit rate, as defined above. Year dummies
controlling for vintage year effects are included but not reported. White
heteroskedasticityconsistent standard errors are shown in italics. ***, **, and * denote
significance at the 1%, 5%, and 10% level.
Conclusion
Networks perform several important roles in venture capital (Dubini and Aldrich, 1991):
allowing access to resources without incurring the costs of vertical integration; acting as
conduits for information; and providing legitimacy to new market entrants (Burt, 1992;
Larson, 1992). Given the important role of syndicates in the financing of new ventures,
understanding the process by which syndicate networks are developed and managed
remains a significant aspect of the entrepreneurial process. In this study, we examine the
performance consequences of a particular form of organizational choice: when market
transactions are characterized by strong relationships and networks. Using VCs syndicate
portfolio company investments in a comprehensive sample of U.K. and Continental
Europe based VCs over the period 1995-2005, we examine the relation between fund
performance and various measures of networking among VCs. We control for various
determinants of VC investment performance, and find that networked VCs realize
significantly better performance. These results suggest that VC firms need to pay close
attention to their relational strategies as networking is likely to add value to firm
operations.
22
References
Aldrich, H. and Zimmer, C. (1986) Entrepreneurship through social networks. In D. L.
Sexton and R. W. Smilor (eds.) The Art and Science of Entrepreneurship (Cambridge,
MA: Ballinger) 2–23.
Barney, J. (1991) Firm resources and sustained competitive advantage, Journal of
Management, 17: 99-120.
Birley, S. (1985) The role of networks in the entrepreneurial process. Journal of Business
Venturing, 1, 107–117.
Bonacich, P. (1987) Power and centrality: A family of measures, American Journal of
Sociology 92, 1170-1182.
Brander, J., Raphael A., and Werner A. (2002) Venture capital syndication: Improved
venture selection versus the value-added hypothesis, Journal of Economics and
Management Strategy 11, 423-452.
British Venture Capital Association. (2002) Report on Investment Activity. London:
BVCA.
Burt, R. S. (1992) The social structure of competition. In N. Nohria and R. G. Eccles
(eds.) Networks and Organizations: Structure, Form, and Action. (Boston, MA: Harvard
Business School Press) 57–91.
Bygrave, W. D. (1988) The structure of the investment networks of venture capital firms,
Journal of Business Venturing 3, 137-158.
23
Chu, P. (1996) Social network models of overseas Chinese entrepreneurship: the
experience of Hong Kong and Canada. Canadian Journal of Administrative Sciences. 13,
358–365.
Cornelli, F. and Goldreich, D. (2001) Bookbuilding and strategic allocation, Journal of
Finance 56, 2337-2369.
Dubini, P. and Aldrich, H. (1991) Personal and extended networks are central to the
entrepreneurial process. Journal of Business Venturing, 6, 305–313.
Elango, B., Fried, V., Hisrich, R. and Polonchek, A. (1995) How venture capital firms
differ. Journal of Business Venturing, 10: 157-179.
Franke, N., Gruber, M., Harhoff, D. and Henkel, J. (2006) What you are is what you like
- similarity biases in venture capitalists’ evaluations of start-up teams. Journal of
Business Venturing, 21(6), 802-826.
Goodstein, J., Gautum, K. and Boeker, W. (1994) The effects of board size and diversity
on strategic change. Strategic Management Journal, 15: 241–250.
Gorman, M., and Sahlman, W. A. (1989) What do venture capitalists do? Journal of
Business Venturing 4, 231-248.
Granovetter, M. S. (1973) The strength of weak ties. American Journal of Sociology, 78,
1360–1380.
Granovetter, M. S. (1985) Economic action and social structure: the problem of
embeddedness. American Journal of Sociology, 91, 481–510.
24
Hochberg, Y., Ljungqvist, A. and Lu, Y. (2006) Whom you know matters: Venture
capital networks and investment performance, Unpublished working paper, New York
University.
Johanson, J. and Mattson, L.G. (1987) Interorganizational relations in industrial systems:
a network approach compared with a transaction-cost approach. International Studies of
Management and Organization 17, 34–48.
Kaplan, S. and Schoar, A. (2005) Private equity returns: Persistence and capital flows,
Journal of Finance 60, 1791-1823.
Kenis, P., and Knoke, D. (2002) How organisational field networks shape
interorganisational tie-formation rates, Academy of Management Review, 27: 275-293
Larson, A. (1992) Network dyads in entrepreneurial settings: a study of the governance
of exchange relationships. Administrative Science Quarterly, 37, 76–104.
Lerner, J. (1994) The syndication of venture capital investments, Financial Management
23, 16-27.
Ljungqvist, A., Marston, F. and Wilhelm, W. J. (2005) Scaling the hierarchy: How and
why investment banks compete for syndicate co-management appointments, Unpublished
working paper, New York University.
Norton, E. and Tenenbaum, B.H. (1993) The effects of venture capitalists’ characteristics
on the structure of a venture capital deal. Journal of Small Business Management, 31(4),
32–41.
Powell, W. W. (1990) Neither market nor hierarchy: network forms of organization. In B.
M. Staw and L.L. Cummings (eds.) Research in Organizational Behavior, 12,
(Greenwich, CT: JAI Press). 295–336.
25
Powell, W. W., Koput, K.W. and Smith-Doerr, L. (1996) Interorganizational
collaboration and the locus of innovation: networks of learning in biotechnology.
Administrative Science Quarterly, 41, 116–145.
Provan, K. G. and Sebastian, J. G. (1998) Networks within networks: Service link
overlap, organizational cliques, and network effectiveness, Academy of Management
Journal, 41: 453-463
Sahlman, W. A. (1990) The structure and governance of venture capital organizations,
Journal of Financial Economics 27, 473-421.
Shulman, N. (1976) Network analysis: a new addition to an old bag of tricks. Acta
Sociological, 19, 307–323.
Steier, L. and Greenwood R. (1995) Venture capitalist relationships in the deal
structuring and postinvestment stages of new firm creation. Journal of Management
Studies, 32, 337–357.
Tichy, N. M. (1981) Networks in organizations. In P. C. Nystrom and William H.
Starbuck (eds.) Handbook of Organizational Design, vol. 2 (Oxford: Oxford University
Press).
Wernerfelt, B. (1984) A resource-based view of the firm, Strategic Management Journal,
5: 171-180.
Wijbenga, F., Postma, T., van Witteloostuijn, A. and Zwart, P. (2003) Strategy and
performance of new ventures: A contingency model of the role and influence of the
venture capitalist’, Venture Capital 5 (3): 231 – 250.
Wilson, R. (1968) The theory of syndicates, Econometrica 36, 199-132.